Feature plotting #466

clarkfitzg · 2015-07-11T00:18:51Z

Not complete, but still ready for some feedback!

Docs are built here:
http://xray.readthedocs.org/en/feature-plotting/plotting.html

There are several TODOs around the docs and the code base. For some of these I'm requesting specific feedback.

Still need to:

Handle NaN's
Informative TypeError for coords that can't be plotted

…rted

shoyer · 2015-07-17T17:10:24Z

doc/plotting.rst

+    a.plot()
+
+It may seem strange that
+the the values on the y axis are decreasing with -0.5 on the top. This is because


shoyer · 2015-07-17T17:20:02Z

xray/core/plotting.py

+    # Lists need to be converted to np.arrays here.
+    if not any(_right_dtype(np.array(x), plottypes) for x in args):
+        raise TypeError('Plotting requires coordinates to be numeric '
+                        'or dates. Try DataArray.reindex() to convert.')


reindex is not a method that converts string labels to numbers:
http://xray.readthedocs.org/en/stable/indexing.html#align-and-reindex

I would suggest DataArray.assign_coords to assign new coordinate labels.

clarkfitzg · 2015-07-18T04:06:11Z

doc/plotting.rst

+    @savefig plotting_example_time.png width=6in
+    noise.plot_line()
+
+TODO- rotate dates printed on x axis.


This isn't plotting well.

jhamman · 2015-07-18T04:19:17Z

I'm going to get around to testing this out tomorrow. I just breezed through the docs and I have one comment. Can we not use jet as the default colormap? I know it is the default colormap in matplotlib but we aren't bound by that. There are dozens of sources out there explaining why jet is a bad choice. I'll give two of the ones that have popped out to me over the years:

There is also a lively Github issue over here: matplotlib/matplotlib#875

I'll try to get a few more comments out tomorrow. So far, this work is really encouraging.

nbren12 · 2015-07-18T04:34:00Z

I agree that jet isn't a good color scheme (this scipy talk is really
good https://www.youtube.com/watch?v=xAoljeRJ3lU), but that is easily
changed by setting cmap='whatever'.

One colorbar issue I find consistently annoying is having to manually
specify vmin and vmax or levels when plotting anomaly fields on a
diverging colorbar with white in the middle. Some sort of
symmetric_colorbar= <middle point value> kwarg would be really nice.

shoyer · 2015-07-18T04:40:38Z

@jhamman I'd be happy to bundle viridis with xray as the default colormap (the source code is here) -- or we could just wait a few months, at which time it will be the matplotlib default.

@nbren12 Seaborn has some heuristics for picking sequential vs diverging colormaps. Possibly worth borrowing: https://github.com/mwaskom/seaborn/blob/v0.6.0/seaborn/matrix.py#L158

My inclination is to merge this on master and iterate in future PRs. There are still a few things to fix, but this is a great start.

jhamman · 2015-07-18T06:02:59Z

yeah, +1 for the viridis colormap.

jhamman · 2015-07-18T16:27:24Z

doc/plotting.rst

+
+    distance = np.linalg.norm(xy, axis=2)
+
+    distance = xray.DataArray(distance, zip(('y', 'x'), (y, x)))


list(zip(..)) for python3 compatibility?

jhamman · 2015-07-18T18:30:16Z

@clarkfitzg - I just spent some time running through your examples and trying to plot some of my own data. Overall - this is really slick and I really liked what you have going here. Below are a few comments I made along the way - take them or leave them.

Consider yincrease=True as the default. Maybe there is a downside to this but I don't see it. I'm pretty sure it is just the convention of imshow.
It may be worth mentioning why you are using imshow over pcolormesh. The limitation of imshow having regularly spaced coordinates will be important to some (myself included). We may, at some point, want to add an option for using pcolormesh. This would support irregular grids without forcing the use of contourf.
More examples with cartopy would be good. Maybe we can put together example page here: http://xray.readthedocs.org/en/latest/examples.html
I continue to have cartopy install problems on osx, even with python2.7 using anaconda. I also had problems on a clean docker image. I know this has nothing to do with this PR but the problems with installing cartopy will limit its applicability here.

shoyer · 2015-07-18T19:18:24Z

Have you tried the ioos conda channel for cartopy? They do a really nice job maintaining builds.

On Sat, Jul 18, 2015 at 11:30 AM, Joe Hamman [email protected]
wrote:

@clarkfitzg - I just spent some time running through your examples and trying to plot some of my own data. Overall - this is really slick and I really liked what you have going here. Below are a few comments I made along the way - take them or leave them.

Consider yincrease=True as the default. Maybe there is a downside to this but I don't see it. I'm pretty sure it is just the convention of imshow.

It may be worth mentioning why you are using imshow over pcolormesh. The limitation of imshow having regularly spaced coordinates will be important to some (myself included). We may, at some point, want to add an option for using pcolormesh. This would support irregular grids without forcing the use of contourf.

More examples with cartopy would be good. Maybe we can put together example page here: http://xray.readthedocs.org/en/latest/examples.html

1. I continue to have cartopy install problems on osx, even with python2.7 using anaconda. I also had problems on a clean docker image. I know this has nothing to do with this PR but the problems with installing cartopy will limit its applicability here.

Reply to this email directly or view it on GitHub:
#466 (comment)

clarkfitzg · 2015-07-19T01:41:18Z

Thanks for all the feedback! I'll change the default colormap and look into the other things mentioned on Monday.

Re Python 3 compatibility with xray.DataArray(d_ylog, zip(('y', 'x'), (y, x))) it would be nice if the DataArray constructor accepted an iterable. Thoughts?

shoyer · 2015-07-19T03:01:36Z

Re Python 3 compatibility with xray.DataArray(d_ylog, zip(('y', 'x'), (y, x))) it would be nice if the DataArray constructor accepted an iterable. Thoughts?

Agreed. Want to add a fix?

clarkfitzg · 2015-07-19T17:17:20Z

Sure. I'll do it separate from this PR.

Feature plotting

shoyer · 2015-07-20T03:20:11Z

Let's follow up on this with some smaller PRs.

clarkfitzg · 2015-07-20T16:07:36Z

@jhamman

The issue with yincrease=True as the default is that actual PNG images when loaded will be 'upside down' when plotted with xray.
Definitely could add support for pcolormesh. Went with imshow since it seems more common.
+1
I recently installed cartopy on a new machine with conda:

conda install -c scitools cartopy

Seems like it used an old version of matplotlib, but it still worked.
Followed their docs: http://scitools.org.uk/cartopy/docs/latest/installing.html

Clark Fitzgerald added 30 commits June 26, 2015 16:29

plotmethod decorator to import matplotlib

3ab7ed1

#185 organizing structure for plotting

e8671f5

begin adding docs for plotting

bec3054

include png output from plot

bd5109f

multivariate normal example

00bed50

fleshing out docs a bit

edcf55c

test for axis label

26b35f1

plotting works in docs for sin example

723be1c

beginning to set up auto docs

cb100f9

test for ylabel

a18ac23

decorator to skip matplotlib tests

2ce0bd9

simple tests passing, working on build

fabbb91

working on making the build work

86b0320

using Agg backend for matplotlib on Travis

eb79aa0

using slices to pick correct indexes for axis labels

a16d0ad

import matplotlib for appveyor windows build

9c1a268

fix python3 bug from itemsview

05c17c8

empty tests with todo for notes on pull requests

1994f9a

add FacetGrid class

2311507

starting on histogram

a768583

imshow

d0b2761

default plot method

9b717af

cleaner plotting tests by assigning to axis object

ad1cab6

reorganizing plotting docs

ad4ec31

testing all plotting funcs take axes arg

13b5a1b

add links to other plotting libs

62981b7

only plot image plots when the data is coordinates are uniform and so…

573723f

…rted

add is_uniform_spaced function to utils

3f13523

add link to holoviews

8864527

plotting docs provide a nice spec now

3bf2375

shoyer reviewed Jul 17, 2015
View reviewed changes

Merge branch 'master' of github.com:xray/xray into feature-plotting

d657bc5

shoyer reviewed Jul 17, 2015
View reviewed changes

incorporate Stephans feedback

9a54bed

clarkfitzg reviewed Jul 18, 2015
View reviewed changes

jhamman reviewed Jul 18, 2015
View reviewed changes

shoyer added a commit that referenced this pull request Jul 20, 2015

Merge pull request #466 from xray/feature-plotting

1faf1b2

Feature plotting

shoyer merged commit 1faf1b2 into master Jul 20, 2015

clarkfitzg mentioned this pull request Jul 20, 2015

DataArray constructor allow iterables #482

Closed

clarkfitzg deleted the feature-plotting branch July 20, 2015 17:47

clarkfitzg mentioned this pull request Aug 18, 2015

Plot methods #185

Closed


		distance = np.linalg.norm(xy, axis=2)

		distance = xray.DataArray(distance, zip(('y', 'x'), (y, x)))

Uh oh!

Feature plotting #466

Feature plotting #466

Uh oh!

Conversation

clarkfitzg commented Jul 11, 2015

Uh oh!

shoyer Jul 17, 2015

Choose a reason for hiding this comment

Uh oh!

clarkfitzg Jul 17, 2015

Choose a reason for hiding this comment

Uh oh!

shoyer Jul 17, 2015

Choose a reason for hiding this comment

Uh oh!

clarkfitzg Jul 17, 2015

Choose a reason for hiding this comment

Uh oh!

clarkfitzg Jul 18, 2015

Choose a reason for hiding this comment

Uh oh!

jhamman commented Jul 18, 2015

Uh oh!

nbren12 commented Jul 18, 2015

Uh oh!

shoyer commented Jul 18, 2015

Uh oh!

jhamman commented Jul 18, 2015

Uh oh!

jhamman Jul 18, 2015

Choose a reason for hiding this comment

Uh oh!

jhamman commented Jul 18, 2015

Uh oh!

shoyer commented Jul 18, 2015

1. I continue to have cartopy install problems on osx, even with python2.7 using anaconda. I also had problems on a clean docker image. I know this has nothing to do with this PR but the problems with installing cartopy will limit its applicability here.

Uh oh!

clarkfitzg commented Jul 19, 2015

Uh oh!

shoyer commented Jul 19, 2015

Uh oh!

clarkfitzg commented Jul 19, 2015

Uh oh!

shoyer commented Jul 20, 2015

Uh oh!

clarkfitzg commented Jul 20, 2015

Uh oh!

Uh oh!